Measurements of cell viability (percentage live versus dead cells) may be used to evaluate: cellular damage due to toxins, the effect of infection, response to drugs, toxicity during early stages of drug development, the life of cancerous cells, the rejection of implanted organs, etc. There is a need to focus on label-free approaches to drug discovery, especially regarding hit identification (focused screening and primary screening/HTS) and lead optimization. Identifying hits generally requires analysis of cell population viability under drug treatment. Early stages of drug discovery rely on screening large numbers of compounds to analyze their toxic effects on cells. These results are based on biochemical assays performed in an ultra high throughput format. First and foremost is the cell viability analysis, i.e., to recognize the live and dead cells in a population of cells treated by various compounds at different concentration levels. Cell death can occur in many ways: apoptosis, autophagy, cornification and necrosis, to name a few. The underlying biochemical causes responsible for events resulting in different kinds of cell deaths are beginning to be better understood. Although cell morphology and significant ultrastructural changes occur during the initiation and progress of cell death, most of the current protocols to identify dead cells and the death mechanism depend on some form of labeling, such as fluorescence. Disadvantages in using labels for viability analysis include difficulty in delivering the dye to the target, keepingthe dye passive, unknown response to high frequency light, unknown response to dye perturbation over a long period, etc. Also, conducting high throughput studies using fluorescent dyes is expensive and requires complex robotics to prepare specimens. The proposed Cell Analytics for Viability Experiments (CAVETM) system would be a significant step forward in achieving label-free, accurate experiments in life sciences and drug discovery for label free viability analysis and hit-analysis. Label-free techniques work in the absence of fluorescence markers; hence, they do not possess the disadvantages of using fluorescent labels. This is also amenable to the study of cells using video microscopy and time-lapse microscopy. The ability to directly measure the morphological and ultra-structural changes of the cell components and relate the changes to cell death phenomena could open up the drug discovery world to the label-free, live primary cell monitoring experiments. In Phase 1, we plan to accomplish complete software development for detecting apoptosis induced cell death using a label-free process. This includes novel mechanisms to segment label-free cells in a dense cell population, measure a large number of features, automatic assay dependent feature selection, and classification of live and dead cells. In Phase 2, viability analysis will be expanded to other forms of cell death induction such as necrosis, autophagy, cornification, etc. and will develop informatics support to predict cell death based on initial cell features. The whole cell viability analysis will also be extended to 2D+T (time) data in the form of time-lapse or video data of live cells. We plan to market the software through our tie ups with microscopy vendors, UC drug discovery center and by directly providing customized services.